The long-term goal is to provide new and improved analytical methods that will enable researchers to find gene markers genetically linked with the putative gene(s) for schizophrenia. This, in turn, will open the way to identifying the schizophrenia gene(s). Once such genes are found, there is hope that knowledge of their structure will lead to a deeper understanding of the disease etiology and, thus, to the possibility of ameliorating or curing schizophrenia. The specific aims consist of developing more efficient methods of linkage analysis than have previously been employed. Also, automated computer- based methods will be developed to allow researchers to decide which families and which family members in a family to ascertain so as to maximize linkage information with a minimum of individuals tested. Nonparametric methods will be extended to allow for phenocopies (affected sib pair methods) and for both recombination and linkage disequilibrium (haplotype relative risk statistic). New likelihood methods of linkage analysis will be derived by focusing attention on a subset of disease phenotypes so as to reduce dependency on model assumptions, by allowing for an oligogenic rather than a monogenic mode of inheritance of schizophrenia, and by considering disease status jointly with biological indicator variables as a multivariate phenotype. The methods will be obtained by statistical investigation and will be implemented in computer programs, partly by modifying existing multi-point computer programs. Programs and methods will be made available to researchers and will also be applied to schizophrenia families collected by other researchers, particularly those of Dr. Baron.